More from the Fairer layoffs Series

A poorly implemented reduction in force (RIF) could result in litigation and liability that is far greater than the amount saved by the layoff.

So before a RIF is implemented, there should be statistical analysis of whether it would result in an adverse impact by gender, race, age or other prohibited factors. If it would, it may be necessary to go back to the drawing board.

"I always insist on a disparate impact analysis before implementing a RIF, and if there is a statistical problem, the RIF gets halted," said Gerald Hathaway, an attorney with Drinker Biddle in New York City. "Regroup and reconsider. I have seen some pretty horrible results that stopped RIFs dead in their tracks."

In one planned RIF, the workforce was one-third white and two-thirds people of color. But a planned layoff of 40 percent of the workforce was not going to affect a single white worker.

In another instance, a manager selected all Asian-Americans in his work group of 50 for a layoff of 15.

Kerry Notestine, an attorney with Littler in Houston, has seen managers cause adverse impact in layoff selections. In one instance, statistical analysis showed a large RIF would have adverse impact against those over age 40. The statistical group narrowed down the reason for the adverse impact to one division among three and then one region among several, finally determining that the entire adverse impact was based on one manager selecting four individuals for layoff, all of whom were over age 70.

"When corporate HR and legal evaluated the selections, they became concerned about the manager and his selections and decided to redo the selections for that store, taking the manager out of the process and instead having the area manager and area HR rep do the selections," Notestine said. "There were several changes in the selections that eliminated the adverse impact."

[SHRM members-only toolkit: Avoiding Adverse Impact in Employment Practices]

Process for Adverse Impact Analysis

Start an adverse impact analysis by listing the employees who would be laid off if the criteria were to be implemented. Then analyze whether certain protected classes are affected more than others by comparing the percentage of the protected class scheduled for layoff with the percentage of the protected class in the employer's entire workforce or the workforce subset selected for evaluation, said Molly Batsch, an attorney with Greensfelder in St. Louis.

For example, if black workers account for 20 percent of an employer's workforce but 80 percent of the employees scheduled to be affected by the RIF, this disparity would present a litigation risk, she said. Often, an employer can make small modifications in the criteria used—such as applying the criteria across a larger pool—that reduces the impact on a particular class. If modifications are not feasible, the employer should make sure that it has a legitimate business purpose for using the criteria at issue. "This purpose should be well-documented and unrelated to any protected classes," Batsch said.

David Cohen, president of DCI Consulting in Washington, D.C., said performance as a criterion is typically business-related, while laying off workers who earn the most may be suspect, as those employees may be the oldest.

Hathaway explained that if adverse impact is due to subjective decision-making or evaluations, the RIF selection criteria should be revisited. An employer should find some objective metric to measure performance. When the objective measures still result in a disparate impact, the employer should try to figure out why that is. "Was one group more efficient because they are using new methods for which they had training while the others did not? I have seen this," he said.

Adverse impact cannot be corrected simply by making selections based on gender or race, because those selections would not be lawful. "So if Sally and Susan and 10 other women get unselected [for layoffs], and Harry and Bill and 10 more men get selected to remove the statistical problem, those men have a slam-dunk discrimination case because they were chosen because they were men," Hathaway said.

Notestine recalled one RIF at a manufacturing facility. The selections resulted in adverse impact against black workers. When notified of the issue, the local HR professional suggested moving several black workers up in the ranking, resulting in fewer black employees being selected for layoff. "When asked the basis for the change, the HR person said to avoid adverse impact," he said. "We rejected the change on this basis and had legal evaluate the selections to be sure the original selection process was valid."

A statistically significant disparity for a protected characteristic such as race, gender or age does not mean the RIF is unlawful, said Stephen Woods, an attorney with Ogletree Deakins in Greenville, S.C. It just means the RIF may be subject to tougher scrutiny if challenged.

[SHRM members-only how-to guide: How to Conduct a Layoff or Reduction in Force]

Use Process to Determine Adverse Impact

The following process determines adverse impact:

  • Calculate the rate of selection for each group. Divide the number of people selected from a group by the number of applicants from that group.
  • Determine which group has the highest selection rate. For positive personnel transactions, such as hiring, the highest rate is the most advantageous. For negative personnel transactions, such as layoffs, the most favored group has the lowest rate.
  • Calculate the impact ratios by comparing the selection rate for each group with that of the highest group. Divide the selection rate for a group by the selection rate for the highest group.
  • Observe whether the selection rate for any group is substantially less (i.e., usually less than four-fifths or 80 percent) than the selection rate for the highest group. If it is, adverse impact is indicated in most circumstances.

In the following example, the impact ratio for termination is calculated for a group of terminated employees.
Calculation example: Impact ratio for termination.
Determine the group with the lowest rate of selection.

​EEO Group
Selected for Termination
Percentage Selected

The group with the lowest selection rate is male, at 7 percent. Next, the impact ratio is calculated. For negative actions, the most favored group's rate is placed in the numerator position.

Impact ratio for termination. Calculate the impact ratio for each group.

​EEO Group
Percentage selected for Termination Divide

Adverse Impact? Yes. The impact ration is less than 80%.

Source: SHRM members-only toolkit: Avoiding Adverse Impact in Employment Practices.

Adverse Impact by Age

It is fairly common to see an adverse impact on a demographic group—most commonly based on age—in a RIF, said David Froiland, an attorney with Ogletree Deakins in Milwaukee.

"The most important question is, why is there an age problem?" Froiland said.

In one RIF for a truck manufacturer, there was an overwhelmingly disparate impact on older workers. "So we asked the question, why?" Froiland recalled.

The company had a department of 25 clay modelers who made prototype parts out of clay so the engineers could visualize the parts. Then, with the dawn of computer-aided design drawing, the company decided to eliminate the clay-modeler department. All 25 clay modelers were older than age 62. When the clay-modeling department was removed from the adverse impact analysis, the statistics looked pristine. "Because we could provide a legitimate and nondiscriminatory reason for the disparity—that is, we could explain the numbers—we concluded that the selections were defensible," Froiland said.

Analyze the protected-age employees for adverse impact by groups: those who are ages 40 to 49, 50 to 59, 60 to 65 and older than 65, recommended Robin Shea, an attorney with Constangy, Brooks, Smith & Prophete in Winston-Salem, N.C.

"Usually we find a lot more adverse impact as employees' ages increase," Shea said. "So in most settings—with the possible exceptions of models and athletes—the 40-to-49s may not be affected too much, the employees in their 50s look a little worse, and the 60-and-overs can look really bad."

She said that employees who are older than 60 often opt for voluntary retirement. "If so, once you remove the volunteers from the analysis, it works out all right."

But she cautioned that if an employer changes its selection criteria once the RIF process is under way, which is common, it should not forget to have HR and legal counsel again review the new criteria and selected employees.

PART 3: Should Employers Notify in Writing or Not?


HR Daily Newsletter

News, trends and analysis, as well as breaking news alerts, to help HR professionals do their jobs better each business day.